Metabolic tumor burden on postsurgical PET/CT predicts survival of patients with gastric cancer

Cancer Imaging

Table 4 Univariable and multivariable Cox regression analysis for associations between clinicopathologic characteristics and OS in patients with positive PET findings (n = 236)

Variables	Number	Univariable		Multivariable
Variables	Number	HR	P values	HR	P values
Age (years)
≥ 50	123 (52.1%)	0.971	0.838
< 50^a	113 (47.9%)	1.000
Gender
Men	114 (48.3%)	0.973	0.851
Women^a	122 (51.7%)	1.000
Pathological type of primary tumor
PAC/TAC	83 (35.2%)	0.832	0.405
PDAC	117 (49.6%)	0.964	0.860
SRC/MAC^a	36 (15.3%)	1.000
Lauren classifications
Intestinal	92 (39.0%)	1.047	0.757
Non-intestinal^a	144 (61.0%)	1.000
T stage of primary tumor
T1	37 (15.7%)	1.046	0.884
T2	75 (31.8%)	1.381	0.239
T3	101 (42.8%)	1.235	0.428
T4^a	23 (9.7%)	1.000
Gastrectomy
Total	115 (48.7%)	0.982	0.899
Subtotal^a	121 (51.3%)	1.000
Lymphovascular invasion
Yes	146 (61.9%)	1.334	0.060
No^a	90 (38.1%)	1.000
Clinical restage
Locoregional recurrence	140 (61.9%)	1.026	0.862
Distant metastasis^a	96 (38.1%)	1.000
Postsurgical adjuvant treatment
Yes	175 (74.2%)	0.906	0.309
No^a	61 (25.8%)	1.000
Postsurgical MTB_WB on PET/CT
SUV_WBmax	236 (100%)	1.066	< 0.001	1.052	0.002
MTV_WB (cm³)	236 (100%)	1.421	< 0.001	1.443	< 0.001
TLG_WB (cm³)	236 (100%)	1.301	0.001	1.332	0.005

Note. OS overall survival, PET positron emission tomography, HR hazard ratio, CI confidence interval, PAC papillary adenocarcinoma, TAC tubular adenocarcinoma, PDAC poorly differentiated adenocarcinoma, SRC signet-ring cell carcinoma, MRC mucinous adenocarcinoma, SUV standardized uptake value, WB whole body, Max maximum, MTV metabolic tumor volume, TLG total lesion glycolysis. ^a, Reference group. Data in parentheses are percentages of corresponding variables

ISSN: 1470-7330